Mingfeng Li

Guilin University of Electronic Technology, Amazon (United States)

Papers

Total Citations

H-Index

About

Mingfeng Li is a robotics researcher whose work focuses on intelligent perception, 3D reconstruction, and autonomous manipulation for industrial and robotic applications. His key research areas include vision-guided welding robotics, dual-arm cooperative control, and single-image 3D generation. Li’s major contributions include developing an improved CPDA method for feature point identification in fillet weld joints, which enhances the precision of vision-guided welding robots—critical for industries like shipbuilding and vehicle manufacturing. His work on dual-arm cooperative task stability addresses challenges in synchronized robotic manipulation, improving safety and efficiency in collaborative environments. More recently, Li has advanced 3D object generation from single unposed RGB images using Gaussian splatting and hybrid diffusion priors, enabling robots to reconstruct complete geometry and texture for precise grasping and scene understanding. While his citation counts are still growing (with his most cited paper at 4 citations), his research represents foundational steps toward more autonomous and dexterous robotic systems. Li’s work sits at the intersection of computer vision, robotics, and AI, with potential impact on autonomous navigation and industrial automation.